Stable LM 2 Zephyr 1.6B
Stable LM 2 Zephyr 1.6B is an earlier instruction-tuned release based on Stable LM 2 1.6B. It follows a Zephyr-style training pipeline with supervised examples and preference optimization.
Research release
Model design and access
Its instruction format is supplied through the tokenizer's chat template. The card identifies public instruction and preference datasets and distinguishes this model from its pretrained base. Zephyr describes the adaptation method and release identity, not a different model provider.Stable LM 2 Zephyr 1.6B official model card
Practical use
Practical guidance. This checkpoint is useful for comparisons of earlier compact alignment methods. Preserve its tokenizer and template with the weights when reproducing behavior. For a new dialogue product, evaluate newer alternatives and the later Stable LM Chat release before selecting it solely because of its small parameter count.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | stablelm-2-zephyr-1_6b |
| Context window | 4,096 tokens Published checkpoint sequence length; prompt and generated text share the window. Stable LM 2 Zephyr 1.6B configuration |
| Language | English.Stable LM 2 Zephyr 1.6B official model card |
| Adaptation | Supervised instruction tuning and DPO.Stable LM 2 Zephyr 1.6B official model card |
| License | Non-Commercial Research Community License.Stable LM 2 Zephyr 1.6B official model card |
Related models
- Stable LM 2 Chat 1.6B
Stability AI · Stable LM
Stable LM 2 Chat 1.6B is an English instruction-tuned version of the 1.6B base.
Launch date not confirmedResearch releaseDownloadable - Stable LM Zephyr 3B
Stability AI · Stable LM
Stable LM Zephyr 3B is an English instruction model adapted from stablelm-3b-4e1t.
Launch date not confirmedDownloadable